JPH07210332A - Disk array device - Google Patents

Abstract

PURPOSE:To move the storage area of stored data at a high speed by transferring the data among disk devices when data transfer inside the same disk device is required, changing the correspondence relation of an actual storage position and a logical address and discriminating the moved data by the logical address for specifying a moving destination. CONSTITUTION:A disk array controller 12 compares the logical address of the data for which movement is instructed with a moving address specified as the moving destination, judges whether or not movement is possible just by the data transfer among the disk devices 14A-14C and transfers the data to a storage area when the movement is possible. When the movement by the data transfer among the disk devices 14A-14C is impossible, arrangements capable of the movement of respective blocks constituting the data are decided and the respective blocks constituting the data are transferred to the decided arrangement destinations. Then, the contents of an address conversion table on a memory are changed so as to let the transferred data be specified when the logical address is specified.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk array device, for example, a disk array device used as an external storage device of a computer.

[0002]

2. Description of the Related Art A disk device used as an external storage device such as a computer has an advantage that it can store a large amount of data as compared with a storage element such as a semiconductor memory. However, since a mechanical operation is required to read data, the response speed may be slow, or the information stored on the disk medium may be lost due to a collision between the head and the disk medium. It also has the drawback. Therefore, in order to improve the response speed and reliability, the data is divided into blocks of a predetermined size and distributed and stored in a plurality of disk devices.

The data storage in such a form can be realized by issuing a command from the computer side to each of a plurality of disk devices, but the load on the computer side becomes heavy. In order to solve this problem, the disk array device is a device that determines the storage procedure on the storage device side.

The structure and operation of a conventional disk array device will be briefly described with reference to FIGS. 9 and 10. As shown in FIG. 9, the disk array device 11 includes a plurality of disk devices 14 ₁ to 14 ₃ that can operate independently.
And a disk array controller 12 for integrally controlling them as one disk device. There are various storage forms used in such a disk array device, and the disk array device is usually configured so that the storage form can be selected.

For example, RAID (Redundant Arrays)
f Inexpensive Disks) The storage form called Level 0
If selected, as shown schematically in Figure 10.
In addition, the data for which a write request has been
Data D of lock unit₁, D ₂, D₃Divided into
Disk device 14₁Through 14₃Distributed in the book
Be impressed. As mentioned above, the disk array device
From the strike side, it can be handled as a single disk device
Configured as That is, during a command from an external device
Does not specify the logical address where the data should be stored.
It is only done, and each data D₁Through D
₃Is not specified. Individual
The storage area to write each data block is a disk area.
The ray controller uses the address information contained in the command.
It is determined based on the information and the selected storage form. Also, RA
In the storage form called ID levels 3 to 5, a predetermined number
For each block, exclusive of data in those blocks
Parity, which is the logical OR, is calculated, and the calculated parity
Is also stored.

The disk array device is a storage device that stores data in such a form. However, since the disk device is used, the data access speed is, of course, the data stored. It changes according to the existing storage area (track position). Therefore, even in the disk array device, there are cases where it is desired to move frequently accessed data to a storage area having a high access speed, as in a normal disk device. In the conventional disk array device, in such a case, the data stored in the disk array device is read once to the host computer,
The read data was rewritten to a different address.

[0007]

[Problems to be Solved by the Invention]

As described above, in the conventional disk array device, since the data movement is performed via the host computer, it takes a lot of time for the processing and a problem that a heavy load is applied to the host computer. was there.

Therefore, an object of the present invention is to provide a disk array device capable of moving a storage area for stored data at high speed.

[0010]

The invention according to claim 1 is
A plurality of disk devices each having a storage unit for storing data in a predetermined size; allocation means for allocating a logical address, which is one-dimensional information, to all the storage units in the plurality of disk devices; Specifying means for specifying the data to be recorded and the logical address of the move destination of the data, and judging means for judging whether or not the movement of the data specified by the specifying means is possible only by the data transfer between the disk devices, When the determination means determines that the data transfer between the disk devices is not possible, the setting means for setting the destination to which the data transfer between the disk devices is possible and the setting means set by the setting means. The transfer unit that transfers the data instructed to move and the storage unit to which the data has been transferred by this transfer unit are designated as the move destination by the specifying unit. As identified by logical addresses, comprising an updating means for updating the allocation content assignment unit.

That is, according to the first aspect of the invention, in order to move the data to the move destination designated by the logical address, when the data transfer within the same disk device is necessary, the actual move destination is changed. However, the data can be moved only by transferring the data between the disk devices. And
By changing the correspondence between the actual storage position and the logical address, the moved data can be identified by the logical address used to specify the moving destination. As a result, all data movement can be performed by data transfer between disk devices, so operations such as writing (to other disk devices) while reading data can be performed, and data movement can be completed at high speed. Will be able to.

According to a second aspect of the present invention, one-dimensional information is stored in a plurality of disk devices each having a storage unit for storing data in a predetermined size, and in all the storage units in the plurality of disk devices. Assigning means for allocating a logical address, designating means for designating a logical address of data to be moved and a storage area range to which the data is to be moved, and a disk device within the storage area range designated by the designating means. Setting means for setting a move destination that can be moved only by data transfer between the transfer means, transfer means for transferring the data instructed to move to the move destination set by the setting means, and data transfer by this transfer means The assigned contents of the assigning means are updated so that the assigned storage unit can be identified by the logical address used for designating the data to be moved in the designating means. It includes an update means for.

That is, according to the second aspect of the invention, a storage area range (for example, a storage area range with a high transfer rate)
By specifying, when moving data to the specified destination, set the move destination so that data transfer within the same disk device is not necessary, and only transfer the data between the disk devices. Make it possible to move. Then, by changing the correspondence between the actual storage position and the logical address, the moved data can be identified by the logical address used to specify the moved data. As a method of specifying the storage area range, for example, a method of dividing the storage area of each disk device into a plurality of areas according to the transfer rate and designating one of the areas can be used. This allows, for example, data files that are frequently accessed to
Work such as moving to a storage area with a high transfer rate can be performed at high speed.

[0014]

EXAMPLES The present invention will be described in detail below with reference to examples.

FIG. 1 shows the configuration of a disk array device according to an embodiment of the present invention. Disk array device 11
Is composed of a disk array controller 12 and three disk devices 14 _A to 14 _C , which are connected to the bus 1
5 are connected to each other. The disk array controller 12 responds to an access request received from an external device such as the host computer 21 by the disk device 14
Is a control circuit for integrally controlling the CPU and is composed of a processor and its peripheral circuits. The storage area 16 shown in each disk device is a storage area provided for storing an address conversion table. Details of the address conversion table will be described later.

The operation of the disk array device of the embodiment will be described below. First, the operation during normal writing will be described. For example, when the disk array controller 12 receives a data write request from the host, based on the address information A _INIT and the size information S specified in the request, the block for which the write request is made is It is determined whether the data is a minute data, and the disk device to write each data block and the address in the disk device are determined. For this determination, an address conversion table that stores the relationship between the logical address used by the external device to instruct the data and the physical address in each disk device is used.

The address conversion table is stored in the address conversion table storage area 16, and is read by the disk array controller 12 when the apparatus is started up and stored in the memory inside the disk array controller. Disk array controller 1 during normal operation
2 determines the operation contents by referring to the address conversion table on the memory.

The disk array controller 12, which has decided the disk device to write each data block and the address in the disk device, divides the received data into blocks and divides each data block into
Write to the determined storage area.

Next, the operation when instructed to move the data will be described. The disk array controller 12 starts the data migration process by being given the information designating the data to be migrated and the information designating the migration destination as the migration command. The disk array device of the embodiment is configured so that, as the information for designating the destination of the movement, either the information clearly designating the address or the information designating only the storage area range can be used. First, the operation when the destination logical address is specified will be described.

FIG. 2 shows a flow of operation of the disk array controller when data movement is instructed. The disk array controller 12 recognizing that the destination address is specified compares the logical address A _{INIT of the} data instructed to be moved with the logical address A _REQ designated as the destination, and the disk device Only by transferring data between them, it is judged whether or not the movement is possible (step S101). Then, if it is possible to move (step S102; Y), the data stored from the logical address A _INIT is transferred to the storage area starting from the logical address A _REQ (step S103), and the processing ends. .

If it is not possible to move the data between the disk devices (step S102; N), an arrangement is made so that each block forming the data can be moved by the data transfer between the disk devices. Is determined (step S104). The details of the arrangement destination setting process will be described later. Then, each block constituting the data designated to be moved is transferred to the allocation destination determined in step S104 (step S105), and when the logical address A _REQ is designated, the data transferred in this step is As specified, the contents of the address translation table on the memory are changed (step S106). The contents of the address conversion table are also written in the address conversion table storage area 16.

Note that step S103 and step S
In 105, the disk array controller 12
Only issues an instruction to start data transfer, and actual data transfer timing control is performed between disk devices.

Referring to FIGS. 3 and 4, step S10
The contents of the judgment in 1 will be described. These figures schematically show the relationship between the storage positions of the data D ₁ to D ₃ designated to be moved and the storage positions of the data D ₁ ′ to D ₃ ′ after the movement. As shown in FIG.
When data that is designated to be moved is transferred to the logical address A _REQ and there is data that does not change in the stored disk device before and after the move, the disk array controller 12 transfers the data between the disk devices. Judge that it is impossible to move. When the disk array controller 1 is in the state as shown in FIG.
In No. 2, it is judged that the data transfer between the disk devices is possible.

The operation contents of the disk array controller shown in FIG. 2 will be further described with reference to FIG.
This figure schematically shows the outline of the address conversion table before and after the execution of the process of step S106, in which (a) shows the contents before the execution and (b) shows the contents after the execution. .

In step S101, the address conversion table in the state (a) is referred to and the logical address A
_{Since the} storage unit specified by _REQ and the storage unit specified by the logical address A _INIT are both in the first disk device, it is determined that data transfer by data transfer between disk devices is impossible. . Therefore, in step S10
In 4, the arrangement destination of each data block is determined.
In determining the placement destination, the placement destination of each block is
Logical address A _REQ and size of data to be moved S (see FIG. 5)
Then, S is determined to be within the storage area designated by “3”) and the allocation destination is on the storage area of the disk device other than the disk device in which the block is stored. Then, in step S105, each data block is moved by data transfer between the disk devices according to the determined content, and in step S106, the address conversion table is rewritten to that shown in (b).

As described above, in the disk array device of the embodiment, the data movement is performed only by the data transfer between the disk devices. Therefore, while reading the data from one disk device, the data is written to the other disk device. It is possible to move, and data movement can be completed at high speed.

In the operation described above, the logical address of the migration destination is designated, but in the disk array device of the embodiment,
The storage area range that is the destination of the movement can be specified. In this case, the move destination is specified by specifying the storage area name classified according to the transfer rate. In the disk array device of the embodiment,
The high transfer rate area, the medium transfer rate area, and the low transfer rate area can be selectively designated.

FIG. 6 shows the operation flow of the disk array controller in this case. The disk array controller first searches the designated transfer rate area for a free area necessary for storing the data (step S201).
If there is no free area (step S202; N), an error message is output to the host computer (step S203), and the process ends.

If there is a free area (step S
201; Y), the migration destination of each block forming the data is determined to be on the storage area of the disk device other than the disk device in which the block is stored (step S).
204) Then, each block forming the data is transferred to the allocation destination determined in step S204 (step S205), and the external device specifies the data transferred in this step at the address used for specifying the data. As described above, the contents of the address conversion table are changed (step S206).

FIG. 7 shows an outline of the processing contents in step S206. Address A as shown in FIG.
If the data written in the _INIT to A _INIT +2 is moved, as shown in FIG. 7 (b), the address A _INIT
~ The device number and physical address corresponding to A _INIT +2 will be rewritten. The device number and physical address (“1” and P
No logical address A _DUM corresponded to _2, etc.) to A _DUM
Naturally, the device number and the physical address designated by the addresses A _{INIT to} A _INIT +2 are given to +2.

In the disk array device of the embodiment,
Although the error message is output when there is not enough free area, the free area may be searched from another transfer rate area and the data may be moved to that area.

In the disk array device of the embodiment,
Since a disk device having a high-performance interface is used, the disk array controller simply outputs a data transfer start instruction to each disk device in step S105 of FIG.
It is possible to execute data transfer. However, when using a disk device in which data cannot be directly exchanged between the disk devices, the data is once read to the memory in the disk array controller 12 and the data in the memory is read to another disk device. An operation such as writing is required, and the effect of setting the transfer destination to another disk device cannot be sufficiently obtained.

In such a case, it is necessary to configure the disk array device as shown in FIG.

Modification

FIG. 8 shows the configuration of a modified disk array device. The modified disk array device 11 includes a disk array controller 12 and a direct transfer control circuit 1.
It is composed of three and three disk devices 14 and a bus switching circuit 17. The direct transfer control circuit 13 is a circuit that actually performs the data transfer that is being executed according to the instruction of the disk array controller 12 in the disk array device of the embodiment, and during the operation, the bus switching circuit 1
7 to control three disk units 14 _A to 14 _C
To form an independently accessible path. That is, when executing data transfer between the disk devices, the direct transfer circuit 13 forms a path with the disk device 14 _A and a path with the disk device 14 _B , for example, and the disk transfer is performed using these two paths. Device 14 _A
While reading out data from the disk device 14 _B,
Write that data.

In the operation of the disk array device of the modified example, the data transfer between the disk devices is performed by the direct transfer control circuit, but the essential operation is the same as the operation of the disk array device of the embodiment. The detailed description is omitted.

In the disk array devices of the embodiments and the modifications, the data migration is started when the data migration instruction is issued. For example, the access frequency of each data is monitored and the low transfer is performed. When data with high access frequency is stored in the rate area, the storage area may be automatically changed.

[0038]

As described above, according to the first aspect of the present invention, since the data movement is performed only by the data transfer between the disk devices, the operation of writing the data (to another disk device) while reading the data can be performed. It becomes possible and the data movement can be completed at high speed.

According to the second aspect of the invention, the work of moving a data file having a high access frequency to a storage area having a high transfer rate can be performed at high speed.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing a configuration of a disk array device according to an embodiment of the present invention.

FIG. 2 is a flowchart showing an operation flow of a data migration process by the disk array system of the embodiment.

FIG. 3 is an explanatory diagram for explaining the data movement in which the disk array device of the embodiment determines that the data movement by the data transfer between the disk devices is impossible.

FIG. 4 is an explanatory diagram for explaining a data movement that the disk array device of the embodiment determines that data movement is possible by data transfer between the disk devices.

FIG. 5 is an explanatory diagram showing how the address conversion table is updated by the processing of FIG.

FIG. 6 is a flowchart showing an operation flow of a data migration process when a migration destination is designated in a range in the disk array system of the embodiment.

FIG. 7 is an explanatory diagram showing how the address conversion table is updated by the processing of FIG.

FIG. 8 is a schematic diagram showing a configuration of a disk array device according to a modified example of the present invention.

FIG. 9 is a schematic diagram showing a configuration of a conventional disk array device.

FIG. 10 is an explanatory diagram outlining a storage operation in a conventional disk array device.

[Explanation of symbols]

11 ... Disk array device, 12 ... Disk array controller, 13 ... Direct transfer control circuit, 14 ... Disk device, 15 ... Bus, 16 ... Address conversion table storage area, 17 ... Bus switching circuit, 21 ... Host computer (external device)

Claims (2)

[Claims] 1. A plurality of disk devices having a storage unit for storing data for each predetermined size, and a logical address, which is one-dimensional information, is assigned to all the storage units in the plurality of disk devices. Allocating means, designating means for designating the data to be moved and the logical address of the destination of the data, and determining whether or not the data designated by this designating means can be moved only by data transfer between disk devices. And a setting means for setting a moving destination that can be moved only by data transfer between disk devices when the determination means determines that it is impossible only by data transfer between disk devices. The transfer means for transferring the data instructed to move to the destination set in step 1 and the storage unit to which the data has been transferred by the transfer means are designated by the designating means. A disk array device comprising: an updating unit that updates the allocation contents of the allocation unit so that the logical address specified as the migration destination is identified. 2. A plurality of disk devices having a storage unit for storing data for each predetermined size, and a logical address, which is one-dimensional information, is assigned to all the storage units in the plurality of disk devices. Allocating means, designating means for designating a logical address of data to be moved and a storage area range to which the data is to be moved, and data transfer between disk devices within the storage area range designated by the designating means. The setting means for setting a movable destination, the transfer means for transferring the data instructed to move to the destination set by the setting means, and the storage unit to which the data is transferred by the transfer means, Update for updating the assignment contents of the assigning unit so that the assigning unit can identify the logical address used for designating the data to be moved. The disk array apparatus characterized by comprising a stage.